자료유형  

 학위논문

Control Number  

0016933629

International Standard Book Number  

9798379565015

Dewey Decimal Classification Number  

004

Main Entry-Personal Name  

Lemmer, Stephan J.

Publication, Distribution, etc. (Imprint  

[S.l.] : University of Michigan., 2023

Publication, Distribution, etc. (Imprint  

Ann Arbor : ProQuest Dissertations & Theses, 2023

Physical Description  

1 online resource(193 p.)

General Note  

Source: Dissertations Abstracts International, Volume: 84-12, Section: B.

General Note  

Advisor: Corso, Jason J.

Dissertation Note  

Thesis (Ph.D.)--University of Michigan, 2023.

Restrictions on Access Note  

This item must not be sold to any third party vendors.

Restrictions on Access Note  

This item must not be added to any third party search indexes.

Summary, Etc.  

요약While deep learning problems are often motivated as enabling technologies for human-computer interaction-a robot, for example, must align natural language referents and sensor readings to operate in a human world-assumptions of these works make them poorly suited to real-world human interaction. Specifically, evaluation typically assumes that humans are oracles that provide semantically correct and unambiguous information, and that all such information is equally useful. While this is enforced in controlled experiments via carefully curated datasets, models operating in the wild will need to compensate for the fact that humans are hazy oracles that may provide information that is incorrect, ambiguous, or misaligned with the features learned by the model. For example: given a choice of three mugs, a robot would not be able to satisfy a request to retrieve the mug, but would likely be able to retrieve the orange mug.A natural question follows: how can we use deep learning models trained via the oracle assumption with hazy humans? We answer this question via a method we call deferred inference, which allows models trained via supervised learning to solicit and integrate additional information from the human when necessary. Deferred inference begins with a method for determining if the model should defer inference and wait until more human-provided information is provided. Past work has generally simplified this into one of two questions: is the human-provided information correct? or is the output correct? However, we find that these approaches are insufficient due to the complex relationship between human inputs, sensor readings, and deep models: low-quality human-provided information may not cause error, while high-quality human-provided information may not correct it. To address the misalignment between input and output error, we introduce Dual-loss Additional Error Regression, or DAER, a method that successfully locates instances where a new human input can reduce error.Although introduction of such an effective deferral function is necessary to optimize the tradeoff between human effort and error, we must additionally consider that the deferral response is also subject to the effects of hazy oracles. For this reason, we must not only consider how to find error caused by human input, but also how to integrate deferral responses and measure the performance of the team. For this, we introduce aggregation functions that allow us to integrate information across multiple inferences and a novel evaluation framework that measures the trade-off between error and additional human effort. Through this evaluation, we show that we can reduce error by up to 48% under a reasonable level of human effort without any changes to training or architecture.Last, we consider how shifting from dataset-based evaluation to an individual human affects deferred inference. Specifically, while crowdsourced datasets work well for rapid implementation and evaluation of deferral and aggregation functions, they do not accurately model human-computer interaction: the mechanisms used to procure high-quality data most likely cause shifts in the distribution, and the failure to track the inputs of individual annotators makes the tacit assumptions that all humans are the same and inputs do not change over time or deferral depth. Through a human-centered experiment, we find that these assumptions are not true: an ideal deferral function must be calibrated for a specific user, users learn the model over time, and the deferral response is likely to be of lower quality than the initial query. Further, we show that despite the shortcomings of crowdsourced data, our deferral method does still reduce error.While deep learned models have been proposed for many applications that require cooperation between humans and computers, deploying models that were trained and evaluated across carefully curated datasets remains a challenge due to the hazy nature of human inputs. In this dissertation, we propose deferred inference as a method for addressing this challenge while respecting the paradigm of supervised training. By demonstrating components of deferred inference on four disparate problems, we provide meaningful insights into its challenges, benefits, and generalizability that motivate and lay the foundation for the eventual deployment of deep-learned human-in-the-loop models in the wild.

Subject Added Entry-Topical Term  

Computer science.

Subject Added Entry-Topical Term  

Robotics.

Index Term-Uncontrolled  

Human-computer interaction

Index Term-Uncontrolled  

Computer vision

Index Term-Uncontrolled  

Deep learning

Index Term-Uncontrolled  

Deferred inference

Added Entry-Corporate Name  

University of Michigan Robotics

Host Item Entry  

Dissertations Abstracts International. 84-12B.

Host Item Entry  

Dissertation Abstract International

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:641002